Arch Womens Ment Health DOI 10.1007/s00737-014-0427-6

ORIGINAL ARTICLE

Antenatal depression: an artefact of sleep disturbance? R. Mellor & S. C. Chua & P. Boyce

Received: 18 July 2013 / Accepted: 7 April 2014 # Springer-Verlag Wien 2014

Abstract Research indicates that poor sleep quality is linked to and may precede depressive symptomatology in pregnancy, complicating screening for either condition. Pregnancy onset may also contribute to the development of sleep-disordered breathing (SDB). For the first time, the link between SDB and depression was examined in pregnancy. A total of 189 pregnant women completed the Edinburgh Postnatal Depression Scale (EPDS), Pittsburgh Sleep Quality Index (PSQI) for sleep quality and the Berlin Questionnaire for SDB. Women were also asked what they felt was the cause of their symptoms. PSQI-assessed poor sleep quality and self-perceived depression were strongly associated with EPDS scores of probable depression (X2 13.39; p<0.001). Berlin-assessed risk of SDB was also associated with probable depression (X2 9.20 p<0.01), though this was attenuated following multivariate analysis. There was a significant relationship between total PSQI score and the tendency for participants to attribute ‘sleep-related causes’ to their low mood (X 2 20.78; p <0.001). This study confirms the link between PSQIassessed poor sleep quality and depressive symptoms in

R. Mellor (*) The University of Sydney, Sydney, NSW, Australia e-mail: [email protected] S. C. Chua Department of Obstetrics and Gynaecology, Westmead Hospital, Sydney, NSW, Australia e-mail: [email protected] P. Boyce Discipline of Psychiatry, Sydney Medical School, The University of Sydney, Sydney, NSW, Australia e-mail: [email protected] P. Boyce Department of Psychiatry, Westmead Hospital, Sydney, NSW, Australia

pregnancy, suggesting the two questionnaires assess the same or overlapping conditions. Although there was a relationship between probable depression and high risk SDB, the effect was attenuated after accounting for other depression risk factors, including body mass index (BMI). Keywords Depressive disorder . Pregnancy . Sleep disorders . Sleep apnea syndromes . Questionnaires

Introduction Prevalence of sleep disturbance and depression in pregnancy There is growing evidence that pregnancy is a time of increased risk for sleep disturbance and depression, both of which are associated with adverse pregnancy outcomes. Worldwide rates of depressive symptomatology in pregnancy as assessed by the Edinburgh Postnatal Depression Scale (EPDS) (Cox et al. 1987) vary from 5 to 39 % (Bowen and Muhajarine 2006; Lau et al. 2011; Hartley et al. 2011; Li et al. 2012; Luke et al. 2009; Melo et al. 2012; Mohammad et al. 2011; Edwards et al. 2008b; Brooks et al. 2009), while the prevalence of diagnosed major depressive disorder is 7.4 to 12.8 % (Banti et al. 2011; Bennett et al. 2004; Gotlib et al. 1989). Differences in prevalence rates depend on the gestational age and socioeconomic status of the subgroup studied as well as the EPDS cut-off used to define probable depression. Sleep disturbance and overall sleep quality are commonly assessed with the Pittsburgh Sleep Quality Index (PSQI) (Buysse et al. 1989); using this tool, rates of poor sleep quality during pregnancy range from 53 to 80 % (Ko et al. 2012; Facco et al. 2010; Ko et al. 2010). Insomnia scales indicate that rates of insomnia are as high as 50 to 62 % (Ko et al. 2012; Kizilirmak et al. 2012; Dorheim et al. 2012), while studies

R. Mellor et al.

looking specifically at restless leg syndrome in pregnancy have found rates of 19 to 30 % (Facco et al. 2010; Ko et al. 2012). In recent years, the nature of the sleep disturbance during pregnancy has been attributed to sleep-disordered breathing (SDB); this refers to a group of disorders characterised by abnormal respiratory patterns and gas exchange during sleep, of which obstructive sleep apnoea (OSA) is the most common (Facco 2011). Use of the Berlin Questionnaire (Netzer et al. 1999) as a screening tool for SDB in pregnant populations indicates rates of 25 to 33 % (Ko et al. 2012; Olivarez et al. 2 0 11 ; H i g g i n s e t a l . 2 0 11 ) , w h i l e t h e r a t e o f polysomnography-diagnosed SDB is 24 % (Facco et al. 2012), compared to less than 5 % in non-pregnant female populations of reproductive age (Olivarez et al. 2011; Young et al. 1993). Snoring and daytime somnolence are key symptoms of SDB that increase during pregnancy; excessive daytime sleepiness rates are 31 to 45 % (Pien et al. 2005; Reutrakul et al. 2011), while snoring rates are 11 to 35 % (Bourjeily et al. 2010; Facco et al. 2010; O’Brien et al. 2012; Ayrim et al. 2011; Loube et al. 1996) compared to 3.5 to 4 % in non-pregnant control groups (Loube et al. 1996; Ayrim et al. 2011). Snoring rates increase by approximately 25 % with the onset of pregnancy (Izci et al. 2006; O’Brien et al. 2012), and new onset snoring (rather than chronic snoring) is related to adverse pregnancy outcomes (O’Brien et al. 2012). However, while snoring is a good predictor of SDB in nonpregnant populations, it may have a poorer predictive value in pregnant populations; a study by Sahin et al. (2008) found that of 35 pregnant women who screened positive for SDB symptoms, only 4 had diagnosable SDB on polysomnography. Adverse pregnancy outcomes of sleep disturbance and depression in pregnancy The importance of the high prevalence of sleep disturbance, SDB and depressive symptomatology during pregnancy lie in their association with adverse outcomes for both mother and foetus. Perinatal complications associated with antenatal depression were examined by two meta-analyses (Grigoriadis et al. 2013; Grote et al. 2010). Both found that depression in pregnancy is associated with a modest but statistically significant increased risk of preterm birth. Grote et al. (2010) also found an association between depression and low birth weight, while Grigoriadis et al. (2013) demonstrated a link between antenatal depression and decreased initiation of breastfeeding. In the long term, antenatal depression may contribute to childhood developmental delay, independent of the occurrence of postnatal depression (Deave et al. 2008; Evans et al. 2012; Hanington et al. 2012). An issue with assessment of childhood development is that much of the research has relied on parental report; a mother with depression may be less able to recognise her child’s abilities and so

rate poorly on development scales. The maternal risks of antenatal depression include greater use of prescription drugs such as hypnotics, anti-emetics and opioid analgesics (Newport et al. 2012; Alder et al. 2011), lower maternal attachment in the postpartum (Perry et al. 2011) and increased risk of anxiety (Skouteris et al. 2009) and postnatal depression (Milgrom et al. 2008; Kim et al. 2008; Goyal et al. 2007). Possible adverse consequences of self-reported snoring or polysomnography-diagnosed SDB in pregnancy include greater risk of gestational hypertensive disorders (Bourjeily et al. 2010; O’Brien et al. 2012; Perez-Chada et al. 2007; Reid et al. 2011; Olivarez et al. 2011), gestational diabetes (Bourjeily et al. 2010; Facco et al. 2012; Reutrakul et al. 2011) and unplanned caesarean sections (Bourjeily et al. 2010). While these studies demonstrate that the risk of adverse pregnancy outcomes is significant even after adjusting for body mass index (BMI), the risk is synergistically increased when obesity and SDB are considered together (Facco et al. 2012; Reid et al. 2011; Qiu et al. 2010). However, the consequences of SDB in pregnancy are not yet entirely clear, and indeed, some research indicates there are no adverse consequences (Loube et al. 1996; Ayrim et al., 2011). PSQIassessed sleep disturbance is associated with preterm labour (Facco et al. 2012; Reutrakul et al. 2011); for every one-point increase on the PSQI in early pregnancy, the odds of preterm birth increase by 25 % (Okun et al. 2011b). However, rather than being an independent risk factor for adverse outcomes, poor sleep may work in conjunction with perceived stress (Okun et al. 2011b) or depression (Okun et al. 2012). Two major mechanisms have been proposed to explain the link between poor sleep, depression and adverse outcomes in pregnancy. The first is an increase in oxidative stress and proinflammatory cytokines. In the general population, increased levels of C-reactive protein (CRP) and interleukin-6 (IL-6) are associated with psychological distress and depression (WiumAndersen et al. 2013; Hiles et al. 2012). Both of these proinflammatory cytokines are increased in women reporting risk factors for antenatal depression, such as stress and poor social support throughout pregnancy (Coussons-Read et al. 2007), and IL-6 is related to higher levels of PSQI-assessed poor sleep quality in pregnant women (Okun et al. 2007). A model developed by Okun et al. (2009) suggests that the rise in proinflammatory cytokines contributes to adverse pregnancy outcomes such as pre-eclampsia, intrauterine growth restriction and preterm birth by interfering with normal vascular remodelling of the placenta. Oxidative stress may also cause endothelial dysfunction and thus contribute to the development of pre-eclampsia (Bourjeily et al. 2010). This may underlie the link between pre-eclampsia and SDB, which involves repetitive episodes of hypoxia and re-oxygenation and thus an increased production of reactive oxygen species (ROS) (Facco 2011). The second proposed mechanism involves the hypothalamic pituitary adrenal (HPA) axis and sympathetic

Antenatal depression: an artefact of sleep disturbance?

nervous system. As reviewed by Field et al. (2010), higher cortisol and noradrenaline levels are associated with both depression and sleep disturbance in pregnancy; cortisol is also one of the suggested mechanisms for the increased risk of preterm birth (Sandman et al. 2006), though a Pakistani study found no relationship (Shaikh et al. 2011).

for ‘probable depression’ will exhibit greater symptoms of SDB as well as poorer overall sleep quality.

Materials and methods Participants

Depression and sleep in pregnancy: what is already known and what this study aims to contribute In non-pregnant populations, there appears to be a bidirectional causal relationship between sleep problems and depression. Insomnia, hypersomnia and fatigue are diagnostic criteria for major depressive disorder (MDD) in the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; (American Psychiatric Association 2000) and insomnia both predicts and is a symptom of depression (Breslau et al. 1996; Jansson-Frojmark and Lindblom 2008; Krystal 2012). In the general population, there is also a strong association between SDB and depression; depressed subjects are five times more likely to experience SDB (Ohayon 2003), and 44.6 to 56 % of those diagnosed with SDB meet the criteria for at least mild depression on screening questionnaires (Kales et al. 1985; McCall et al. 2006). In recent years, a number of studies have explored the nature of the relationship between sleep and depressive symptomatology in pregnancy, though, to our knowledge, none have focussed specifically on SDB as the cause of sleep disturbance. Use of the PSQI to screen for sleep problems in pregnancy has revealed a significant association between poor sleep quality and depressive symptomatology (Ko et al. 2010; Skouteris et al. 2008; Kamysheva et al. 2010; Jomeen and Martin 2007). Poor sleep quality appears to precede depressive symptoms (Skouteris et al. 2008), though this relationship may be influenced by pregnancy-related physical symptoms (Kamysheva et al. 2010) and perceived stress (Ko et al. 2010). Studies looking at other aspects of sleep disturbance in pregnancy show an association between insomnia and depressive symptomatology (Dorheim et al. 2012; Kizilirmak et al. 2012) as well as longer daytime naps, shorter overnight sleep duration and more fragmented sleep in pregnant women with diagnosed depression (Okun et al. 2011a). The significance of the association between sleep problems and depression appears to be greatest early in pregnancy; by 30 weeks of gestation, sleep is disturbed regardless of depression status (Okun et al. 2011a). While these studies all confirm that general sleep disturbance in pregnancy is linked to and may precede depressive symptomatology, little is known about the exact form this sleep disturbance takes. The aim of this study is to further characterise the link between sleep disturbance and depression in pregnancy, focusing specifically on the contribution of SDB. We hypothesised that participants who screen positive

Following university and local health district ethics approval, 189 pregnant women were recruited to participate in a study entitled ‘Sleep patterns and feelings of wellbeing in pregnancy’. All women attending the hospital antenatal clinics between November 2012 and March 2013 were invited to participate via a poster on the wall or discussion with receptionist and midwifery staff. During this period, an estimated 1,350 women were seen in all clinics combined; up to 16 % were in high-risk clinics. Procedure Participants were asked to self-complete a questionnaire and return it to the study team by either placing it in a locked box in the waiting room or via the reply-paid envelope provided. Questionnaires were non-identifiable and participants were informed that by completing and returning the questionnaire they were consenting to participate in the study. In the first part of the questionnaire, women were asked to report a number of baseline characteristics. Sociodemographic characteristics included age, marital status, education level, employment status, living circumstances and level of support from family and friends. Due to the multicultural nature of the area in which participants were recruited, women were also asked if they spoke a language other than English at home and whether or not they had immigrated to Australia within the last 5 years. Obstetric characteristics were gestation in weeks, gravidity and number of dependent children. Participants were also asked about their history of depression and SDB. Participants then completed two sleep quality measures and one depression measure, finishing with questions on their perception of why they felt the way they did and what could improve it. Sleep measures used were the Pittsburgh Sleep Quality Index (PSQI) (Buysse et al. 1989) and the modified Berlin Questionnaire (Netzer et al. 1999). The depression measure used was the Edinburgh Postnatal Depression Scale (EDPS) (Cox et al. 1987). Instruments The PSQI evaluates general sleep quality and disturbances over a 1-month period. It is made up of 19 items, which generate seven components: overall sleep quality, sleep latency, duration of sleep, sleep efficiency, sleep disturbance,

R. Mellor et al.

needing medication to sleep and daytime dysfunction due to sleepiness. The participant is given a global sleep quality score from 0 to 21 and is defined as having poor sleep quality with a score greater than 5. The PSQI has internal consistency and reliability in pregnant populations (Jomeen and Martin 2007), and the Cronbach α value (Cronbach and Warrington 1951) for this study was 0.74, indicating acceptable internal consistency for this sample. As previous studies found (Jomeen and Martin 2007), this improved to 0.76 when the item ‘need medication to sleep’ was removed. The Berlin Questionnaire specifically assesses the risk of sleep-disordered breathing (SDB). It is made up of 11 questions, divided into three categories; the first category is made up of 5 questions on snoring, the second contains 4 questions on daytime sleepiness and the third contains 2 questions on high blood pressure history and body mass index (which is positive if greater than 30). Each category is scored as positive or negative according to the participant’s answers, and the participants are labelled as high risk for SDB if two or more of the categories are positive. The Berlin Questionnaire has internal consistency and reliability (Sharma et al. 2006), though no tool has yet been validated for SDB screening in pregnant populations. The Cronbach α values for this sample were 0.69 for the items in category 1 and 0.52 for category 2. As previous studies found (Netzer et al. 1999), when the item on ‘drowsiness behind the wheel’ was removed, this increased to 0.62 for category 2. However, these Cronbach α values are less than 0.7, indicating reduced internal consistency of the Berlin Questionnaire for this particular sample. The EPDS is made up of 10 items that assess how the participant felt over the past 7 days only. It is scored out of 30 and a score greater than or equal to 13 indicates probable antenatal depression. This cut-off has greater than 95 % sensitivity and specificity for depression. The EPDS has internal consistency and reliability (Fernandes et al. 2011; Rubertsson et al. 2011; Su et al. 2007; Tandon et al. 2012; Murray and Cox 1990), and the Cronbach’s α for this sample was 0.88. Statistical methods The explanatory variables were examined using descriptive statistics. Medians and interquartile ranges were used for nonnormally distributed continuous variables. Categorical variables were analysed using frequencies and proportions. The three outcome variables are the total scores of the PSQI, EPDS and Berlin questionnaires. The total PSQI and EPDS scores are continuous variables with non-normal distributions, while the total Berlin score is categorical. Each of these scores was further grouped into low- and high-risk groups, and these data used in subsequent chi-squared analyses. The Fisher’s exact test was used where expected cell frequencies were too low. Significance was measured to the p<0.05 level. Multivariate analyses were performed on all

three outcome variables. As the EPDS and PSQI residuals are normally distributed, stepwise linear regressions were performed (with the entry threshold set at p=0.05 and the exit threshold set at p=0.10 for all variables). Logistical regression was used to determine the predictors of SDB.

Results Sample characteristics The baseline characteristics of the sample are shown in Table 1. The sample consisted of 189 pregnant women ranging in age from 17 to 44 years and in gestation from 12 to 39 weeks. A third were primiparous, half had one or more dependent children at home and most (95 %) stated that they felt supported by family and friends. Two thirds spoke a language other than English at home, of which the majority spoke either Mandarin or Cantonese. Twenty percent had immigrated to Australia within the last 5 years, of which half came from China. The majority had long-term partners, were university-educated with a Bachelor’s degree or higher and were employed in paid work at the time of completing the questionnaire. Two participants had a history of sleep apnoea, 12 % had a history of diagnosed depression and 21 % had a BMI of greater than 30. Due to the nature of recruitment, we do not know how many women declined participation and consequently whether or not they were similar to those who did participate. Outcome measures The median EPDS score was 7 (IQR 3–11), and 16 % of participants were defined as having probable depression with scores greater than or equal to 13. The median PSQI score was 6 (IQR 4–8), and 52 % were categorised as having poor sleep quality with scores greater than 5. According to the Berlin Questionnaire, 19 % had a high probability of SDB and 31.9 % were snorers. Six percent of the EPDS questionnaires were incomplete, compared to 21.7 % of the PSQI and 24.3 % of the Berlin questionnaires. Only participants who had completed all three questionnaires were included in the analyses involving overall EPDS, PSQI and Berlin scores; however, where individual factors were examined, participants were re-included as appropriate. Poor sleep quality was significantly associated with probable depression (OR 10.61; 95 % CI 2.32 to 46.86; p<0.001), as was high probability SDB (OR 4.37; 95 % CI 1.58 to 11.82; p=0.005). Other factors associated with probable depression include history of depression, BMI greater than 30, unpaid employment and lack of support from family and friends (Table 2). High BMI and history of depression were both also associated with poor sleep quality, while high BMI, increasing

Antenatal depression: an artefact of sleep disturbance? Table 1 Baseline and clinical characteristics of the sample (N=189) Number (%) Socio-demographic characteristics Age (years) Partnered Language other than English None Chinese Other Immigration in last 5 years No From China From elsewhere Education University degree Diploma or certificate High school Did not complete high school Current employment status Full-time Part-time/casual Unpaid work Current living arrangements Own house Renting Live with family/friends Feel supported by family/friends Obstetric characteristics Gestation (weeks) Gravidity 1 2 ≥3 No. of dependent children 0 1 ≥2 Health characteristics BMI ≤30 >30 History of sleep apnoea History of depression Self-report measures Sleep measures Total PSQI score Poor sleep quality High-probability SDB Depression measures Total EPDS score Probable depression

Median (IQR)

31 (28–34) 183 (97) 71 (40) 42 (24) 66 (37) 147 (80) 21 (11) 16 (9) 121 (64) 35 (19) 25 (13) 8 (4) 94 (51) 52 (28) 37 (20) 128 (69) 42 (23) 15 (8) 174 (95) 24 (19–29) 2 (1–3) 67 (35) 61 (32) 61 (32) 1 (0–1) 89 (48) 70 (37) 28 (15)

136 (79) 36 (21) 2 (1) 22 (12)

Multivariate analysis

6 (4–8) 77 (52) 27 (19) 7 (3–11) 28 (16)

gestation and Australian residency during the last 5 years were associated with SDB (Table 2). Self-reported high blood pressure was not a significant predictor of SDB risk; however, this may be due to a lack of power as only six participants reported having high blood pressure. Interestingly, there was no significant relationship between SDB and sleep quality. When participants were asked whether or not they considered themselves to be depressed, 11 % answered that they did. As shown in Table 3, these participants were significantly more likely to be depressed according to the EPDS (OR 25.82; 95 % CI 8.32 to 80.13; p<0.001), have poor sleep quality according to the PSQI (OR 7.62; 95 % CI 1.62 to 34.56; p=0.003) and be at high risk of SDB according to the Berlin Q (OR 4.65; 95 % CI 1.49 to 14.21; p=0.01). Of those who screened positive for probable depression on the EPDS, 48 % did not consider themselves depressed. Participants were also asked to what they attributed their mood and what could improve their mood. The most common causes for a low mood included specific non-pregnancyrelated stressors, such as work stress (23 %), general pregnancy related changes, including ‘hormones’ (17 %), and specific pregnancy-related stressors, such as fear for the foetus’ wellbeing (15 %) and fatigue (15 %). Three participants (2 %) gave diagnosed anxiety or depression as the reason for their low mood, 11 % were unsure and 16 % stated they had no issues with their mood. Factors that could improve mood were varied; the majority (57 %) gave specific improvements that were unrelated to sleep or pregnancy, and 11 % stated that more or better sleep would help while 6 % stated that their mood would be improved by delivery of a healthy baby. Causes and improvements were further grouped into ‘sleeprelated’ and ‘non-sleep-related’ for bivariate and multivariate analyses. Within the EPDS defined probable depression group, significantly, more women gave non-sleep-related causes for their low mood (p=0.006), while those with PSQI defined ‘poor sleep quality’ gave sleep-related causes for their low mood (p<0.0001). The majority of women (88 %) who attributed ‘sleep-related causes’ for their low mood had poor sleep quality according to the PSQI. There were no significant associations between probability of SDB and self-perceived causes of low mood or possible improvements to mood.

Stepwise linear regression was conducted to identify independent determinants of EPDS score. The factors that had a significant correlation with EPDS score on bivariate analysis were included in the model, along with recent immigration, which has been identified as a risk factor for antenatal depression (Miszkurka et al. 2012; Giardinelli et al. 2012; Lau et al. 2011). Though single motherhood is also an established risk factor (La Porte et al. 2012; Melo et al. 2012), this study comprised only six un-partnered women and so lacked the

R. Mellor et al. Table 2 Relationship between baseline participant characteristics and each of the three outcome measures: depression (EPDS score ≥13), poor sleep (PSQI score >5) and sleep-disordered breathing (Berlin Q ≥2 positive categories) Depressed

Poor sleep quality

High-probability SDB

% (n/N)

Χ2

% (n/N)

Χ2

% (n/N)

Χ2

13.1 (8/61) 13.8 (9/65) 21.2 (11/52)

1.64

52.8 (28/53) 50.0 (26/52) 52.4 (22/42)

0.10

17.6 (9/51) 17.0 (9/53) 22.5 (9/40)

0.52

15.1 (26/172) 33.3 (2/6)

1.45a

50.7 (72/142) 80.0 (4/5)

1.66a

19.6 (27/138) 0.0 (0/6)

1.45a

20.6 (14/68) 14.6 (6/41) 11.3 (7/62)

2.16

52.6 (30/57) 42.4 (14/33) 54.2 (26/48)

1.22

17.3 (9/52) 9.1 (3/33) 30.0 (15/50)

5.82

15.1 (21/139) 20.0 (4/20) 21.4 (3/14)

0.62a

54.6 (65/119) 40.0 (6/15) 33.3 (3/9)

2.45a

22.5 (25/111) 0.0 (0/15) 7.1 (1/14)

5.78a*

14.8 (17/115) 18.8 (6/32) 16.1 (5/31)

0.30

50.0 (48/96) 52.2 (12/23) 57.1 (16/28)

0.45

18.1 (17/94) 23.1 (6/26) 16.7 (4/24)

0.42a

11.4 (10/88) 10.2 (5/49) 34.3 (12/35)

11.51**

51.5 (35/68) 54.5 (24/44) 50.0 (15/30)

0.17

16.2 (11/68) 18.6 (8/43) 29.6 (8/27)

2.26

17.9 (22/123) 10.8 (4/37) 7.1 (1/14)

1.90

54.9 (56/102) 37.9 (11/29) 61.5 (8/13)

3.12

21.6 (21/97) 15.6 (5/32) 9.1 (1/11)

1.36

14.0 (23/164) 62.5 (5/8)

13.15a**

49.6 (67/135) 83.3 (5/6)

2.61a

17.7 (23/130) 37.5 (3/8)

1.93a

17.7 (11/62) 14.3 (9/63) 15.1 (8/53)

0.31

49.1 (27/55) 43.8 (21/48) 63.6 (28/44)

3.88

6.3 (3/48) 22.4 (11/49) 27.7 (13/47)

7.81* P trendb 0.008

9.4 (6/64) 19.0 (11/58) 19.6 (11/56)

3.06

46.2 (24/52) 55.3 (26/47) 54.2 (26/48)

1.00

18.8 (9/48) 21.3 (10/47) 16.3 (8/49)

0.39

11.8 (10/85) 20.0 (13/65) 19.2 (5/26)

2.12

50.0 (36/72) 48.0 (24/50) 64.0 (16/25)

1.87

19.4 (13/67) 16.7 (9/54) 19.0 (4/21)

0.16

13.3 (17/128) 27.0 (10/37)

3.96* OR 2.42

48.1 (51/106) 72.4 (21/29)

5.40* OR 2.83

8.0 (9/113) 58.1 (18/31)

40.08*** OR 16.00

Socio-demographic characteristics Age (years) ≤29 30–33 ≥34 Marital status Partner No partner Language English only Chinese Other Immigration in last 5 years None From China From elsewhere Education level Bachelor or higher Diploma Secondary or less Employment Full-time, paid Part-time, paid Unpaid Living arrangement Own house Renting Live with family/friends Feel supported Yes No Obstetric characteristics Gestation (weeks) ≤20 21–28 ≥29 Gravidity 1 2 ≥3 Dependent children 0 1 ≥2 Maternal health characteristics BMI ≤30 >30

Antenatal depression: an artefact of sleep disturbance? Table 2 (continued) Depressed

High BP No or unsure Yes Hx sleep apnoea No Yes Hx depression No Yes Outcome measures EPDS: probable depression No (<13) Yes (≥13) PSQI: sleep quality Good Poor Berlin Q: probability SDB Low High

Poor sleep quality

High-probability SDB

% (n/N)

Χ2

% (n/N)

Χ2

% (n/N)

Χ2

15.4 (26/169) 0.0 (0/7)

1.26a

50.7 (71/140) 66.7 (4/6)

0.58a

17.4 (24/138) 50.0 (3/6)

4.01a OR 4.75

15.4 (27/175) 50.0 (1/2)

1.78a

51.7 (75/145) 100.0 (1/1)

0.93a

18.9 (27/143) 0.0 (0/1)

0.23a

12.3 (19/155) 40.9 (9/22)

11.88a** OR 4.96

45.2 (56/124) 94.7 (18/19)

16.22*** OR 21.86

16.7 (20/120) 38.9 (7/18)

4.91a OR 3.18

45.9 (56/122) 90.0 (18/20)

13.39*** OR 10.61

14.7 (17/116) 42.9 (9/21)

9.20a** OR 4.37

13.5 (7/52) 24.6 (15/61)

2.22 OR 2.10

2.9 (2/68) 24.3 (18/74)

13.39*** OR 10.61

10.8 (12/111) 34.6 (9/26)

9.20a** OR 4.37

50.5 (46/91) 68.2 (15/22)

2.22 OR 2.10

Odds ratio, in italic a

Fisher’s exact test

b

The linear-by-linear association: used to test for a trend in the frequency of the outcome across an ordered exposure variable

*p<0.05; **p<0.01; ***p<0.001

Table 3 Relationship between self-perception and each of the three outcome measures: depression (EPDS score ≥13), poor sleep (PSQI score >5) and sleep disordered breathing (Berlin Q ≥2 positive categories) Depressed

Self-perceived depression No Yes Causes for low mood No mood issues Sleep-related Non-sleep-related Improvements for mood No mood issues Sleep-related Non-sleep-related a

Fisher’s exact test

b

Odds ratio, in italic

Poor sleep quality

High-probability SDB

% (n/N)

Χ2

% (n/N)

Χ2

% (n/N)

Χ2

7.7 (12/155) 68.4 (13/19)

50.65a*** ORb 25.82

46.0 (58/126) 86.7 (13/15)

8.85** OR 7.62

15.8 (19/120) 46.7 (7/15)

8.15a* OR 4.65

0.0 (0/22) 10.5 (2/19) 25.6 (22/86)

8.50a**

11.8 (2/17) 87.5 (14/16) 61.1 (44/72)

20.78***

15.4 (2/13) 30.8 (4/13) 22.9 (16/70)

0.87a

0.0 (0/21) 16.7 (2/12) 27.5 (22/80)

7.69a*

42.1 (8/19) 60.0 (6/10) 57.6 (38/66)

1.55

6.3 (1/16) 37.5 (3/8) 29.7 (19/64)

4.23a

*p<0.05; **p<0.01; ***p<0.001

R. Mellor et al.

power to include this in the model. Two participants had a history of SDB; this was not associated with any of the outcome measures on bivariate analysis, and given the lack of power, they were excluded from multivariate analyses. Of the 10 factors examined, 4 remained significant in the final model (see Table 4): poor sleep quality based on total PSQI score, self-perception of depression and self-perceived causes for low mood (both sleep and non-sleep related). Overall, the final model explained 50 % of the variance in EPDS scores (p<0.001). Stepwise linear regression was also performed to identify independent determinants of PSQI score. The five factors that had a significant correlation with PSQI on bivariate analysis were examined, along with risk of SDB. The three that remained significant in the final model were EPDS score, a history of depression and self-perceived sleep-related causes for low mood (Table 5). Overall, the final model explained 36 % of the variance in PSQI scores (p<0.001). Independent determinants of SDB were examined using logistic regression. Factors that were significantly associated with SDB on bivariate analysis were included, along with poor sleep quality (PSQI score). BMI greater than 30 was the only significant contributor to SDB risk in the final model (OR 32.57; 95 % CI 9.11 to 116.40; p<0.001), which explained 48 % of the variance in SDB (X2 36.04; p<0.001).

Discussion The high prevalence and associated adverse consequences of sleep disturbance and depressive symptomatology in pregnancy highlight the need to further explore these relationships. In this sample, 16 % met the criteria for probable depression and 52 % for poor sleep quality, which is within the ranges quoted in the literature (see “Introduction”). Nineteen percent of the sample had probable SDB, which is lower than the 25 to 33 % found in other studies (Ko et al. 2012; Olivarez et al. 2011; Higgins et al. 2011); however, snoring rates were within range at 31.9 % (Bourjeily et al. 2010; Facco et al. 2010; O’Brien Table 4 Final model from the stepwise multiple regression for the prediction of probable depression Variable

Standardized beta

t

Self-perception of depression Overall sleep quality score (PSQI) Self-perceived causes for low mood

0.38 0.23

4.37*** 2.62*

0.25 0.46

2.41* 4.55***

Sleep-related Non-sleep-related

Overall model: adjusted R square 0.502; F 22.45; p<0.0001 *p<0.05; ***p<0.001

Table 5 Final model from the stepwise multiple regression for the prediction of poor sleep quality Variable

Standardized beta

t

History of depression Overall depression score (EPDS) Self-perceived causes for mood Sleep-related

0.31 0.43

3.57*** 5.04***

0.23

2.73**

Overall model: adjusted R square 0.352; F 18.03; p<0.0001 **p<0.01; ***p<0.001

et al. 2012; Ayrim et al. 2011; Loube et al. 1996). One of the possible reasons for the lower SDB rate is the multicultural nature and relatively high immigration rate of the study location, as lack of immigration was associated with SDB in this sample. This association may be due to differences in BMI and body habitus between people of Caucasian and Asian descent, as many of the recently immigrated people in this sample belong to the latter. This study confirmed a number of the factors that have previously been associated with depression in pregnancy, namely, lack of employment (Pottinger et al. 2009; Rubertsson et al. 2005), self-perceived lack of support (Lau et al. 2011) and history of depression (Edwards et al. 2008a; Giardinelli et al. 2012; Hromi-Fiedler et al. 2011; Pottinger et al. 2009; Lancaster et al. 2010). Lack of a partner was not associated with significantly increased risk (La Porte et al. 2012; Melo et al. 2012); however, this may be due to a lack of power as there were only six un-partnered women in the sample. PSQI-assessed poor sleep quality was strongly associated with EPDS-assessed probable depression on both bivariate and multivariate analyses, confirming the link described in previous studies (Ko et al. 2010; Skouteris et al. 2008; Kamysheva et al. 2010; Jomeen and Martin 2007). These results may be reflective of a strong link between antenatal depression and poor sleep quality in pregnancy. Alternatively, they may indicate overlap in the factor design of the PSQI and EPDS or the two questionnaires may be identifying the same symptom set and attributing it to different sources, in which case the depressive symptoms seen may be the result of poor sleep rather than diagnosable depression. For example, lack of energy is a key symptom of depression but may also be a consequence of poor sleep. In some cases, it is possible that the effect is summative and both sleep disturbance and diagnosable depression exist within the same individual, augmenting the rating of both conditions. In the current study, SDB was examined for the first time as the potential source of sleep disturbance that is associated with depressive symptomatology in pregnancy. Although there was a relationship between probable depression and high risk SDB, the effect was attenuated after accounting for other risk factors, including BMI. However, no questionnaire has yet

Antenatal depression: an artefact of sleep disturbance?

been validated for SDB screening in pregnancy, and the Cronbach α value for the Berlin Questionnaire in this sample indicates that a more reliable measure is needed before firm conclusions can be made. The link between high BMI and all three outcome measures indicates that the issue of obesity in pregnancy also merits further investigation. Women who scored high in any or all of the three outcome variables were significantly more likely to consider themselves depressed. This suggests that not only are rates of depressive symptomatology and sleep disturbance high in pregnancy, but they also negatively impact on the women affected. When asked about causes for low mood, 87.5 % of those who gave sleep-related causes did have poor sleep quality according to the PSQI. Giving a sleep-related cause for low mood remained a significant predictor of PSQIassessed poor sleep quality on multivariate analysis, whereas giving any cause (sleep- or non-sleep-related) for low mood was an independent predictor of EPDS score. One of the limitations of the current study is the tools used, which were self-report, symptom-based screening measures for depression and sleep, rather than the respective gold standards of structured clinical interview and polysomnography. One study focusing on the immediate postpartum period found that subjective, self-report measures of poor sleep like the PSQI are more highly correlated with mood disturbances than objective measures (Bei et al. 2010). Thus, use of the PSQI may have overestimated the strength of the association between poor sleep and depressive symptomatology. Given the multicultural nature of the sample, another limitation is the English language questionnaire that was used. While the women had to speak enough English to read a poster and interact with the receptionist or midwife in order to receive a questionnaire, misinterpretation of individual questions may have impacted on the results obtained. The two sleep questionnaires were longer and may have been harder for someone whose first language was not English to understand, which may explain the lower rates of completion of these questionnaires as compared to the EPDS. Finally, the extent to which the results are generalizable to the pregnant population as a whole is limited by the risk of selection bias, linked to the poor response rate. The women who chose to participate in the study may have been those with sleep or mental health issues. Despite these limitations, the results of this study do have implications for screening, managing and preventing sleep disorders and depression during pregnancy. They indicate that universal perinatal screening is warranted, but a tool that differentiates between the symptomatology of depression and that of sleep disturbance is still needed. Such a screening tool would need to be implemented in each pregnancy, as it was found that approximately 10 % of women screen discordantly across successive pregnancies (La Porte et al. 2012). Treatment options for depressive and sleep symptomatology in pregnancy are varied. In women with hypertension and

chronic snoring, nasal CPAP used in the first 8 weeks of pregnancy is associated with better blood pressure control and improved pregnancy outcomes, though none of the women in this study had diagnosed SDB (Poyares et al. 2007). Physical activity (Demissie et al. 2011) and yoga or massage therapy (Field et al. 2012) from partners may help depressive symptoms, while cognitive behavioural therapy and interventions focused on the marital relationship may also be of use (Clatworthy 2012). In addition to these strategies, both pregnant women (Rowan et al. 2012) and healthcare professionals (Gawley et al. 2011; Jones et al. 2012) need to be educated on the benefits of screening and treating sleep and mood disturbances in pregnancy. This study focused on screening, diagnosis and one potential mechanism of sleep disturbance, namely, SDB. Future investigation into other forms of sleep disturbance, such as restless legs syndrome, would be of benefit. Furthermore, in order to disentangle the relationship between disturbed sleep and depression, a longitudinal study of non-depressed women using objective measures of sleep disturbance (such as actigraphy), serial assessments of depressive symptoms (using observer-rated scales) and robust diagnostic instruments is required.

Conclusions Self-report poor sleep quality is strongly associated with depressive symptomatology in pregnancy, though current screening tools may be assessing the same or highly overlapping conditions. Symptoms of SDB are also associated with depressive symptomatology, though the effect was attenuated after controlling for other factors, including BMI. Acknowledgments We are extremely grateful to all the pregnant women who participated in this study and to the midwives and receptionist staff for their help in recruiting them. Thanks also to Prof. Richard Trethowan, A/Prof. Barbara Griffin and Mr. Andrew Mellor for their statistical advice.

References Alder J, Fink N, Urech C, Hosli I, Bitzer J (2011) Identification of antenatal depression in obstetric care. Arch Gynecol Obstet 284(6):1403–1409 American Psychiatric Association (2000) Diagnostic and statistical manual of mental disorders, 4th edn. Text Revision American Psychiatric Association, Washington DC Ayrim A, Keskin EA, Ozol D, Onaran Y, Yiidirim Z, Kafali H (2011) Influence of self-reported snoring and witnessed sleep apnea on gestational hypertension and fetal outcome in pregnancy. Arch Gynecol Obstet 283(2):195–199 Banti S, Mauri M, Oppo A, Borri C, Rambelli C, Ramacciotti D, Montagnani MS, Camilleri V, Cortopassi S, Rucci P, Cassano GB (2011) From the third month of pregnancy to 1 year postpartum.

R. Mellor et al. Prevalence, incidence, recurrence, and new onset of depression. Results from the perinatal depression-research & screening unit study. Compr Psychiatry 52(4):343–351 Bei B, Milgrom J, Ericksen J, Trinder J (2010) Subjective perception of sleep, but not its objective quality, is associated with immediate postpartum mood disturbances in healthy women. Sleep 33(4): 531–538 Bennett HA, Einarson A, Taddio A, Koren G, Einarson TR (2004) Prevalence of depression during pregnancy: systematic review. Obstet Gynecol 103(4):698–709 Bourjeily G, Raker CA, Chalhoub M, Miller MA (2010) Pregnancy and fetal outcomes of symptoms of sleep-disordered breathing. Eur Respir J 36(4):849–855 Bowen A, Muhajarine N (2006) Prevalence of antenatal depression in women enrolled in an outreach program in Canada. J Obstet Gynecol Neonatal Nurs 35(4):491–498 Breslau N, Roth T, Rosenthal L, Andreski P (1996) Sleep disturbance and psychiatric disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry 39(6):411–418 Brooks J, Nathan E, Speelman C, Swalm D, Jacques A, Doherty D (2009) Tailoring screening protocols for perinatal depression: prevalence of high risk across obstetric services in Western Australia. Arch Women Ment Health 12(2):105–112 Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ (1989) The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 28(2):193–213 Clatworthy J (2012) The effectiveness of antenatal interventions to prevent postnatal depression in high-risk women. J Affect Disord 137(1–3):25–34 Coussons-Read ME, Okun ML, Nettles CD (2007) Psychosocial stress increases inflammatory markers and alters cytokine production across pregnancy. Brain Behav Immun 21(3):343–350 Cox JL, Holden JM, Sagovsky R (1987) Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry 150:782–786 Cronbach L, Warrington W (1951) Coefficient alpha and internal structure of tests. Psychometrika 16:167–188 Deave T, Heron J, Evans J, Emond A (2008) The impact of maternal depression in pregnancy on early child development. BJOG 115(8): 1043–1051 Demissie Z, Siega-Riz AM, Evenson KR, Herring AH, Dole N, Gaynes BN (2011) Physical activity and depressive symptoms among pregnant women: the PIN3 study. Arch Women Ment Health 14(2):145– 157 Dorheim SK, Bjorvatn B, Eberhard-Gran M (2012) Insomnia and depressive symptoms in late pregnancy: a population-based study. Behav Sleep Med 10(3):152–166 Edwards B, Galletly C, Semmler-Booth T, Dekker G (2008a) Antenatal psychosocial risk factors and depression among women living in socioeconomically disadvantaged suburbs in Adelaide, South Australia. Aust N Z J Psychiatry 42(1):45–50 Edwards B, Galletly C, Semmler-Booth T, Dekker G (2008b) Does antenatal screening for psychosocial risk factors predict postnatal depression? A follow-up study of 154 women in Adelaide, South Australia. Aust N Z J Psychiatry 42(1):51–55 Evans J, Melotti R, Heron J, Ramchandani P, Wiles N, Murray L, Stein A (2012) The timing of maternal depressive symptoms and child cognitive development: a longitudinal study. J Child Psychol Psychiatry 53(6):632–640 Facco FL (2011) Sleep-disordered breathing and pregnancy. Semin Perinatol 35(6):335–339 Facco FL, Kramer J, Ho KH, Zee PC, Grobman WA (2010) Sleep disturbances in pregnancy. Obstet Gynecol 115(1):77–83 Facco FL, Liu CS, Cabello AA, Kick A, Grobman WA, Zee PC (2012) Sleep-disordered breathing: a risk factor for adverse pregnancy outcomes? Am J Perinatol 29(4):277–282

Fernandes MC, Srinivasan K, Stein AL, Menezes G, Sumithra RS, Ramchandani PG (2011) Assessing prenatal depression in the rural developing world: a comparison of two screening measures. Arch Women Ment Health 14(3):209–216 Field T, Diego M, Hernandez-Reif M (2010) Prenatal depression effects and interventions: a review. Infant Behav Dev 33(4):409–418 Field T, Diego M, Hernandez-Reif M, Medina L, Delgado J, Hernandez A (2012) Yoga and massage therapy reduce prenatal depression and prematurity. J Bodyw Mov Ther 16(2):204–209 Gawley L, Einarson A, Bowen A (2011) Stigma and attitudes towards antenatal depression and antidepressant use during pregnancy in healthcare students. Adv Health Sci Educ Theory Pract 16(5):669– 679 Giardinelli L, Innocenti A, Benni L, Stefanini MC, Lino G, Lunardi C, Svelto V, Afshar S, Bovani R, Castellini G, Faravelli C (2012) Depression and anxiety in perinatal period: prevalence and risk factors in an Italian sample. Arch Women Ment Health 15(1):21–30 Gotlib IH, Whiffen VE, Mount JH, Milne K, Cordy NI (1989) Prevalence rates and demographic characteristics associated with depression in pregnancy and the postpartum. J Consult Clin Psychol 57(2):269– 274 Goyal D, Gay CL, Lee KA (2007) Patterns of sleep disruption and depressive symptoms in new mothers. J Perinat Neonatal Nurs 21(2):123–129 Grigoriadis S, Vonderporten E, Mamisashvili L, Tomlinson G, Dennis C, Koren G, Steiner M, Mousmanis P, Cheung A, Radford K, Martinovic J, Ross L (2013) The impact of maternal depression during pregnancy on perinatal outcomes: a systematic review and meta-analysis. J Clin Psychiatry 74(4):e321–e341 Grote NK, Bridge JA, Gavin AR, Melville JL, Iyengar S, Katon WJ (2010) A meta-analysis of depression during pregnancy and the risk of preterm birth, low birth weight, and intrauterine growth restriction. Arch Gen Psychiatry 67(10):1012–1024 Hanington L, Heron J, Stein A, Ramchandani P (2012) Parental depression and child outcomes—is marital conflict the missing link? Child Care Health Dev 38(4):520–529 Hartley M, Tomlinson M, Greco E, Comulada WS, Stewart J, le Roux I, Mbewu N, Rotheram-Borus MJ (2011) Depressed mood in pregnancy: prevalence and correlates in two Cape Town peri-urban settlements. Reprod Health 8:9 Higgins N, Leong E, Park CS, Facco FL, McCarthy RJ, Wong CA (2011) The Berlin Questionnaire for assessment of sleep disordered breathing risk in parturients and non-pregnant women. Int J Obstet Anesth 20(1):22–25 Hiles SA, Baker AL, de Malmanche T, Attia J (2012) A meta-analysis of differences in IL-6 and IL-10 between people with and without depression: exploring the causes of heterogeneity. Brain Behav Immun 26(7):1180–1188 Hromi-Fiedler A, Bermudez-Millan A, Segura-Perez S, Perez-Escamilla R (2011) Household food insecurity is associated with depressive symptoms among low-income pregnant Latinas. Matern Child Nutr 7(4):421–430 Izci B, Vennelle M, Liston WA, Dundas KC, Calder AA, Douglas NJ (2006) Sleep-disordered breathing and upper airway size in pregnancy and post-partum. Eur Respir J 27(2):321–327 Jansson-Frojmark M, Lindblom K (2008) A bidirectional relationship between anxiety and depression, and insomnia? A prospective study in the general population. J Psychosom Res 64(4):443–449 Jomeen J, Martin CR (2007) Assessment and relationship of sleep quality to depression in early pregnancy. J Reprod Infant Psychol 25(1):87– 99 Jones CJ, Creedy DK, Gamble JA (2012) Australian midwives’ awareness and management of antenatal and postpartum depression. Women’s Birth 25(1):23–28 Kales A, Caldwell AB, Cadieux RJ, Vela-Bueno A, Ruch LG, Mayes SD (1985) Severe obstructive sleep apnea—II: Associated

Antenatal depression: an artefact of sleep disturbance? psychopathology and psychosocial consequences. J Chronic Disord 38(5):427–434 Kamysheva E, Skouteris H, Wertheim EH, Paxton SJ, Milgrom J (2010) A prospective investigation of the relationships among sleep quality, physical symptoms, and depressive symptoms during pregnancy. J Affect Disord 123(1–3):317–320 Kim JJ, Gordon TEJ, La Porte LM, Adams M, Kuendig JM, Silver RK (2008) The utility of maternal depression screening in the third trimester. Am J Obstet Gynecol 199(5):509, e501-505 Kizilirmak A, Timur S, Kartal B (2012) Insomnia in pregnancy and factors related to insomnia. ScientificWorldJournal 2012:197093 Ko H, Shin J, Kim MY, Kim YH, Lee J, Kil KC, Moon HB, Lee G, Sa-Jin K, Kim BI (2012) Sleep disturbances in Korean pregnant and postpartum women. J Psychosom Obstet Gynaecol 33(2):85–90 Ko S-H, Chang S-C, Chen C-H (2010) A comparative study of sleep quality between pregnant and nonpregnant Taiwanese women. J Nurs Scholarsh 42(1):23–30 Krystal AD (2012) Psychiatric disorders and sleep. Neurol Clin 30(4): 1389–1413 La Porte LM, Kim JJ, Adams M, Du H, Silver RK (2012) The pattern of depression screening results across successive pregnancies. Am J Obstet Gynecol 206(3):261, e261-264 Lancaster CA, Gold KJ, Flynn HA, Yoo H, Marcus SM, Davis MM (2010) Risk factors for depressive symptoms during pregnancy: a systematic review. Am J Obstet Gynecol 202(1):5–14 Lau Y, Yin L, Wang Y (2011) Antenatal depressive symptomatology, family conflict and social support among Chengdu Chinese women. Matern Child Health J 15(8):1416–1426 Li J, Mao J, Du Y, Morris JL, Gong G, Xiong X (2012) Healthrelated quality of life among pregnant women with and without depression in Hubei, China. Matern Child Health J 16(7):1355–1363 Loube DI, Poceta JS, Morales MC, Peacock MD, Mitler MM (1996) Selfreported snoring in pregnancy. Association with fetal outcome. Chest 109(4):885–889 Luke S, Salihu HM, Alio AP, Mbah AK, Jeffers D, Berry EL, Mishkit VR (2009) Risk factors for major antenatal depression among lowincome African American women. J Women’s Health (Larchmt) 18(11):1841–1846 McCall WV, Harding D, O’Donovan C (2006) Correlates of depressive symptoms in patients with obstructive sleep apnea. J Clin Sleep Med 2(4):424–426 Melo EF Jr, Cecatti JG, Pacagnella RC, Leite DFB, Vulcani DE, Makuch MY (2012) The prevalence of perinatal depression and its associated factors in two different settings in Brazil. J Affect Disord 136(3): 1204–1208 Milgrom J, Gemmill AW, Bilszta JL, Hayes B, Barnett B, Brooks J, Ericksen J, Ellwood D, Buist A (2008) Antenatal risk factors for postnatal depression: a large prospective study. J Affect Disord 108(1–2):147–157 Miszkurka M, Goulet L, Zunzunegui MV (2012) Antenatal depressive symptoms among Canadian-born and immigrant women in Quebec: differential exposure and vulnerability to contextual risk factors. Soc Psychiatry Psychiatr Epidemiol 47(10): 1639–1648 Mohammad KI, Gamble J, Creedy DK (2011) Prevalence and factors associated with the development of antenatal and postnatal depression among Jordanian women. Midwifery 27(6):e238–e245 Murray D, Cox J (1990) Screening for depression during preg- nancy with the Edinburgh Depression Scale (EPDS). J Reprod Infant Psychol 8:99–107 Netzer NC, Stoohs RA, Netzer CM, Clark K, Strohl KP (1999) Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern Med 131(7):485–491

Newport DJ, Ji S, Long Q, Knight BT, Zach EB, Smith EN, Morris NJ, Stowe ZN (2012) Maternal depression and anxiety differentially impact fetal exposures during pregnancy. J Clin Psychiatry 73(2): 247–251 O’Brien LM, Bullough AS, Owusu JT, Tremblay KA, Brincat CA, Chames MC, Kalbfleisch JD, Chervin RD (2012) Pregnancy-onset habitual snoring, gestational hypertension, and preeclampsia: prospective cohort study. Am J Obstet Gynecol 207(6):487, e481-489 Ohayon MM (2003) The effects of breathing-related sleep disorders on mood disturbances in the general population. J Clin Psychiatry 64(10):1195–1200, quiz, 1274–1196 Okun ML, Hall M, Coussons-Read ME (2007) Sleep disturbances increase interleukin-6 production during pregnancy: implications for pregnancy complications. Reprod Sci 14(6):560–567 Okun ML, Kiewra K, Luther JF, Wisniewski SR, Wisner KL (2011a) Sleep disturbances in depressed and nondepressed pregnant women. Depress Anxiety 28(8):676–685 Okun ML, Luther JF, Wisniewski SR, Sit D, Prairie BA, Wisner KL (2012) Disturbed sleep, a novel risk factor for preterm birth? J Women’s Health (Larchmt) 21(1):54–60 Okun ML, Roberts JM, Marsland AL, Hall M (2009) How disturbed sleep may be a risk factor for adverse pregnancy outcomes. Obstet Gynecol Surv 64(4):273–280 Okun ML, Schetter CD, Glynn LM (2011b) Poor sleep quality is associated with preterm birth. Sleep 34(11):1493–1498 Olivarez SA, Ferres M, Antony K, Mattewal A, Maheshwari B, SangiHaghpeykar H, Aagaard-Tillery K (2011) Obstructive sleep apnea screening in pregnancy, perinatal outcomes, and impact of maternal obesity. Am J Perinatol 28(8):651–658 Perez-Chada D, Videla AJ, O’Flaherty ME, Majul C, Catalini AM, Caballer CA, Franklin KA (2007) Snoring, witnessed sleep apnoeas and pregnancy-induced hypertension. Acta Obstet Gynecol Scand 86(7):788–792 Perry DF, Ettinger AK, Mendelson T, Le H-N (2011) Prenatal depression predicts postpartum maternal attachment in low-income Latina mothers with infants. Infant Behav 34(2):339–350 Pien GW, Fife D, Pack AI, Nkwuo JE, Schwab RJ (2005) Changes in symptoms of sleep-disordered breathing during pregnancy. Sleep 28(10):1299–1305 Pottinger AM, Trotman-Edwards H, Younger N (2009) Detecting depression during pregnancy and associated lifestyle practices and concerns among women in a hospital-based obstetric clinic in Jamaica. Gen Hosp Psychiatry 31(3):254–261 Poyares D, Guilleminault C, Hachul H, Fujita L, Takaoka S, Tufik S, Sass N (2007) Pre-eclampsia and nasal CPAP: part 2. Hypertension during pregnancy, chronic snoring, and early nasal CPAP intervention. Sleep Med 9(1):15–21 Qiu C, Enquobahrie D, Frederick IO, Abetew D, Williams MA (2010) Glucose intolerance and gestational diabetes risk in relation to sleep duration and snoring during pregnancy: a pilot study. BMC Womens Health 10:17 Reid J, Skomro R, Cotton D, Ward H, Olatunbosun F, Gjevre J, Guilleminault C (2011) Pregnant women with gestational hypertension may have a high frequency of sleep disordered breathing. Sleep 34(8):1033–1038 Reutrakul S, Zaidi N, Wroblewski K, Kay HH, Ismail M, Ehrmann DA, Van Cauter E (2011) Sleep disturbances and their relationship to glucose tolerance in pregnancy. Diabetes Care 34(11):2454–2457 Rowan P, Greisinger A, Brehm B, Smith F, McReynolds E (2012) Outcomes from implementing systematic antepartum depression screening in obstetrics. Arch Women Ment Health 15(2):115–120 Rubertsson C, Borjesson K, Berglund A, Josefsson A, Sydsjo G (2011) The Swedish validation of Edinburgh Postnatal Depression Scale (EPDS) during pregnancy. Nord J Psychiatry 65(6):414–418 Rubertsson C, Wickberg B, Gustavsson P, Radestad I (2005) Depressive symptoms in early pregnancy, two months and one year postpartum-

R. Mellor et al. prevalence and psychosocial risk factors in a national Swedish sample. Arch Women Ment Health 8(2):97–104 Sahin FK, Koken G, Cosar E, Saylan F, Fidan F, Yilmazer M, Unlu M (2008) Obstructive sleep apnea in pregnancy and fetal outcome. Int J Gynaecol Obstet 100(2):141–146 Sandman CA, Glynn L, Schetter CD, Wadhwa P, Garite T, Chicz-DeMet A, Hobel C (2006) Elevated maternal cortisol early in pregnancy predicts third trimester levels of placental corticotropin releasing hormone (CRH): priming the placental clock. Peptides 27(6): 1457–1463 Shaikh K, Premji SS, Rose MS, Kazi A, Khowaja S, Tough S (2011) The association between parity, infant gender, higher level of paternal education and preterm birth in Pakistan: a cohort study. BMC Pregnancy Childbirth 11:88 Sharma SK, Vasudev C, Sinha S, Banga A, Pandey RM, Handa KK (2006) Validation of the modified Berlin questionnaire to identify patients at risk for the obstructive sleep apnoea syndrome. Indian J Med Res 124(3):281–290 Skouteris H, Germano C, Wertheim EH, Paxton SJ, Milgrom J (2008) Sleep quality and depression during pregnancy: a prospective study. J Sleep Res 17(2):217–220

Skouteris H, Wertheim EH, Rallis S, Milgrom J, Paxton SJ (2009) Depression and anxiety through pregnancy and the early postpartum: an examination of prospective relationships. J Affect Disord 113(3):303–308 Su K-P, Chiu T-H, Huang C-L, Ho M, Lee C-C, Wu P-L, Lin C-Y, Liau C-H, Liao C-C, Chiu W-C, Pariante CM (2007) Different cutoff points for different trimesters? The use of Edinburgh Postnatal Depression Scale and Beck Depression Inventory to screen for depression in pregnant Taiwanese women. Gen Hosp Psychiatry 29(5):436–441 Tandon SD, Cluxton-Keller F, Leis J, Le H-N, Perry DF (2012) A comparison of three screening tools to identify perinatal depression among low-income African American women. J Affect Disord 136(1–2):155–162 Wium-Andersen MK, Orsted DD, Nielsen SF, Nordestgaard BG (2013) Elevated C-reactive protein levels, psychological distress, and depression in 73, 131 individuals. JAMA Psychiatry 70(2): 176–184 Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S (1993) The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 328(17):1230–1235